The purpose of this study is to determine the dose-response effects of methylphenidate treatment on laboratory-measured impulsive behavior of two groups of adolescents, those with either childhood- or adolescent-onset Conduct Disorder comorbid with Attention Deficit/Hyperactivity Disorder (CD/ADHD). CD/ADHD has been shown to be distinctly different in presentation, prognosis, and outcome than either CD or ADHD alone. While stimulants are recommended as the first line of medication treatment for individuals with comorbid CD/ADHD, little is known about the underlying behavioral mechanisms affected by these medications, or how age of CD onset relates to these mechanisms. Researchers have hypothesized that stimulants work by affecting behavioral processes related to impulsivity, but this has not been directly tested. Because impulsivity is a complex construct, a multidimensional approach is necessary to adequately characterize the effects of stimulants on this behavior. For this study, we will recruit three groups of adolescents: childhood-onset CD/ADHD (Early-Onset) and adolescent-onset CD/ADHD (Late-Onset), as well as a group of healthy adolescents (Controls). Baseline (unmedicated) performance on three laboratory- measured components of impulsivity will be compared, including component processes of response initiation, response inhibition, and consequence sensitivity. After baseline performance is evaluated, tolerability for methylphenidate treatment will be confirmed prior to beginning the 3 weeks of randomized methylphenidate treatments using a double-blind, crossover design using three doses of methylphenidate (i.e., 0, 20, and 40 mg). During this 3-week period, behavior will be assessed at home and at school by adolescents, parents, and teachers to test real-world treatment response. At the end of each randomized treatment week, adolescents will be assessed using laboratory behavioral measures of impulsivity to determine the dose-dependent effects of methylphenidate. The primary aims and hypotheses are consistent with our preliminary data are: (1) to determine how performance on each of the major components of laboratory-measured impulsivity differs by age of CD onset; (2) to determine the dose-dependent effects of methylphenidate on performance on each of the major components of impulsivity; and (3) to determine how age of CD onset is differentially related to improvements produced by methylphenidate in both laboratory-measured impulsivity and self/observer ratings of behavior. The innovation and clinical significance of this study is that it will determine how stimulant treatment differentially affects groups of adolescents with either childhood- or adolescent-onset CD comorbid with ADHD; both traditional self/observer ratings and newer laboratory-based strategies for measuring discrete components of impulsive behavior will be used. Relative to individuals having either CD or ADHD alone, the comorbid CD/ADHD group has an exceptionally poor developmental prognosis; therefore, the knowledge gained from this investigation will inform future treatment strategies for this difficult-to-treat population.